Qwen: Qwen3 30B A3B Instruct 2507
qwen/qwen3-30b-a3b-instruct-2507
Cheapest provider
$0.09 / 1M
SiliconFlow
Fastest provider (p95)
135 tok/s
Nebius
Intelligence (estimated)
58.0
Family + generation + popularity + price tier
Per-provider performance
Latency / throughput / uptime measured across providers over the last 30 minutes of live traffic. Atlas’s router weighs these per call (with the eval-gate signal) when picking a variant for Standard and Batch tiers.
| Provider | Quant | Input $/1M | Output $/1M | Latency p50 / p95 | Throughput p50 / p95 | Uptime 30m | Success |
|---|---|---|---|---|---|---|---|
| SiliconFlow | q8· fp8 | $0.0900 | $0.3000 | 3335ms / 6235ms | 18 / 49.4 tok/s | 43.47% | 49.0% |
| AtlasCloud | q8· fp8 | $0.1000 | $0.3000 | 2413ms / 5475ms | 34 / 59 tok/s | 99.76% | 99.9% |
| Nebius | q8· fp8 | $0.1000 | $0.3000 | 726ms / 5456ms | 41 / 135 tok/s | 71.33% | 73.0% |
| WandB | full· bf16 | $0.1000 | $0.3000 | 393ms / 3129ms | 89 / 116 tok/s | 100.00% | 99.7% |
| Alibaba | undisclosed | $0.1300 | $0.5200 | 413ms / 1072ms | 61 / 115 tok/s | 99.58% | 99.6% |
| Venice | undisclosed | $0.1900 | $0.6900 | 1048ms / 1814ms | 18 / 46 tok/s | — | 33.3% |
“—” means live telemetry hasn’t accumulated enough recent traffic for that endpoint. “undisclosed” means the provider serves the model but doesn’t expose the quantization label (typically running fp8 / int8 internally).
Intelligence estimate
No public benchmark numbers indexed for this model yet, so the leaderboard score is derived from the catalogue data we sync: model family standing, generation, and live usage rank — applied identically to open- and closed-weight models. The heuristic ceiling sits below confirmed-benchmark frontier models so curated rankings stay clearly on top.
- ·qwen sm family base
Want a curated score? File a benchmark report and we’ll add it to the next sync.
How Atlas routes Qwen: Qwen3 30B A3B Instruct 2507
- Realtime tier — direct passthrough at the upstream’s native precision. Best for hard latency / quality guarantees.
- Standard tier — Atlas picks the cheapest provider variant whose quantization has stayed green on your operation’s eval gates. For Qwen: Qwen3 30B A3B Instruct 2507 that’s currently SiliconFlow at $0.09/1M.
- Batch tier — async, biggest discount. Roadmapped to use provider batch APIs (OpenAI / Anthropic) where available and queued spot capacity for open-weight workloads.